US10289433B2 - Domain specific language for encoding assistant dialog - Google Patents

Abstract

Systems and processes for generating output dialogs for virtual assistants are provided. An output dialog can be generated from multiple output segments that can each include a string of one or more characters or words. The contents of an output segment can be selected from multiple possible outputs based on a predetermined order, conditional logic, or a random selection. The output segments can be concatenated to form the output dialog. In one example, a dialog generation file that defines the possible outputs for each output segment, an ordering of the output segments within the output dialog, and format for the output dialog can be used to generate the output dialog. The dialog generation file can include any number of functional blocks, which can each output an output segment, that can be arranged hierarchically and in a particular order to generate a desired output dialog.

Description

CROSS-REFERENCE TO RELATED APPLICATION

This application claims priority from U.S. Provisional Ser. No. 62/005,886, filed on May 30, 2014, entitled DOMAIN SPECIFIC LANGUAGE FOR ENCODING ASSISTANT DIALOG, which is hereby incorporated by reference in its entirety for all purposes.

FIELD

This relates generally to virtual assistants and, more specifically, to generating output dialogs for virtual assistants.

BACKGROUND

Intelligent automated assistants (or virtual assistants) provide an intuitive interface between users and electronic devices. These assistants can allow users to interact with devices or systems using natural language in spoken and/or text forms. For example, a user can access the services of an electronic device by providing a spoken user input in natural language form to a virtual assistant associated with the electronic device. The virtual assistant can perform natural language processing on the spoken user input to infer the user's intent and operationalize the user's intent into tasks. The tasks can then be performed by executing one or more functions of the electronic device, and a relevant output can be returned to the user in natural language form.

To generate the output in natural language form, conventional virtual assistants require that a developer specify the content and format of every possible natural language output (e.g., in full-sentence form), as well as the specific conditions that must be met for each output to be returned to the user. For example, a virtual assistant can be programmed to output the string “Your search returned <results.number> results” in response to a search query yielding multiple results, can be programmed to output the string “Your search returned one result” in response to a search query yielding one result, and can be programmed to output the string “Your search returned no results” in response to a search query yielding no results.

While conventional virtual assistants programmed in this way can produce relevant outputs in natural language form, the process of specifying the content and format of every possible natural language output can be time-consuming and can produce repetitive results.

SUMMARY

Systems and processes for operating a virtual assistant are disclosed. One example process can include accessing a textual representation of user speech, determining a response to the textual representation of user speech, generating a plurality of output segments based on the determined response to the textual representation of user speech, wherein each of the plurality of output segments comprises one or more words, and generating an output dialog based on the plurality of output segments and an order associated with the plurality of output segments.

BRIEF DESCRIPTION OF THE DRAWINGS

FIG. 1 illustrates an exemplary environment in which a virtual assistant can operate according to various examples.

FIG. 2 illustrates an exemplary user device according to various examples.

FIG. 3 illustrates a visual representation of an example dialog generation file that can be used to generate an output dialog of a virtual assistant according to various examples.

FIG. 4 illustrates a visual representation of an example output generation file and the output dialog that can be generated using the output generation file.

FIG. 5 illustrates an exemplary process for generating an output dialog for a virtual assistant according to various examples.

FIG. 6 illustrates an example conversation interface that can be displayed on a user device according to various examples.

FIG. 7 illustrates a functional block diagram of an electronic device configured to generate an output dialog for a virtual assistant according to various examples.

DETAILED DESCRIPTION

In the following description of examples, reference is made to the accompanying drawings in which it is shown by way of illustration specific examples that can be practiced. It is to be understood that other examples can be used and structural changes can be made without departing from the scope of the various examples.

For descriptive purposes, throughout this disclosure, software, software modules, software blocks, software objects, and the like may be described as performing various functions. One of ordinary skill in the art, however, will recognize that software may not actively perform any function and instead may include instructions that are executable on a computer processor. As such, although software may be described herein as performing a function, it should be appreciated that a computer processor or other computing device may typically perform those functions attributed herein to software modules, blocks, or objects by executing computer instructions provided by the software modules, blocks, or objects.

This relates to systems and processes for generating output dialogs for virtual assistants. An output dialog can be generated from multiple output segments that can each include a string of one or more characters or words. The contents of an output segment can be selected from multiple possible outputs based on a predetermined order, conditional logic, or a random selection. The output segments can then be concatenated to form the output dialog. In some examples, a dialog generation file that defines the possible outputs for each output segment, an ordering of the output segments within the output dialog, and format for the output dialog to be displayed or otherwise presented to a user can be used to generate the output dialog. The dialog generation file can include any number of functional blocks, which can each output an output segment, that can be arranged hierarchically and in a particular order to generate a desired output dialog.

System Overview

FIG. 1 illustrates exemplary system100 for implementing a virtual assistant according to various examples. The terms “virtual assistant,” “digital assistant,” “intelligent automated assistant,” or “automatic digital assistant” can refer to any information processing system that interprets natural language input in spoken and/or textual form to infer user intent, and performs actions based on the inferred user intent. For example, to act on an inferred user intent, the system can perform one or more of the following: identifying a task flow with steps and parameters designed to accomplish the inferred user intent; inputting specific requirements from the inferred user intent into the task flow; executing the task flow by invoking programs, methods, services, APIs, or the like; and generating output responses to the user in an audible (e.g., speech) and/or visual form.

A virtual assistant can be capable of accepting a user request at least partially in the form of a natural language command, request, statement, narrative, and/or inquiry. Typically, the user request seeks either an informational answer or performance of a task by the virtual assistant. A satisfactory response to the user request can include provision of the requested informational answer, performance of the requested task, or a combination of the two. For example, a user can ask the virtual assistant a question, such as “Where am I right now?” Based on the user's current location, the virtual assistant can answer, “You are in Central Park.” The user can also request the performance of a task, for example, “Please remind me to call Mom at 4 p.m. today.” In response, the virtual assistant can acknowledge the request and then create an appropriate reminder item in the user's electronic schedule. During the performance of a requested task, the virtual assistant can sometimes interact with the user in a continuous dialogue involving multiple exchanges of information over an extended period of time. There are numerous other ways of interacting with a virtual assistant to request information or performance of various tasks. In addition to providing verbal responses and taking programmed actions, the virtual assistant can also provide responses in other visual or audio forms (e.g., as text, alerts, music, videos, animations, etc.).

An example of a virtual assistant is described in Applicants' U.S. Utility application Ser. No. 12/987,982 for “Intelligent Automated Assistant,” filed Jan. 10, 2011, the entire disclosure of which is incorporated herein by reference.

As shown inFIG. 1, in some examples, a virtual assistant can be implemented according to a client-server model. The virtual assistant can include a client-side portion executed on auser device102, and a server-side portion executed on aserver system110.User device102 can include any electronic device, such as a mobile phone, tablet computer, portable media player, desktop computer, laptop computer, PDA, television, television set-top box, wearable electronic device, or the like, and can communicate withserver system110 through one ormore networks108, which can include the Internet, an intranet, or any other wired or wireless public or private network. The client-side portion executed onuser device102 can provide client-side functionalities, such as user-facing input and output processing and communications withserver system110.Server system110 can provide server-side functionalities for any number of clients residing on arespective user device102.

Server system110 can include one or more virtualassistant servers114 that can include a client-facing I/O interface122, one ormore processing modules118, data andmodel storage120, and an I/O interface toexternal services116. The client-facing I/O interface122 can facilitate the client-facing input and output processing for virtualassistant server114. The one ormore processing modules118 can utilize data andmodel storage120 to determine the user's intent based on natural language input, and perform task execution based on inferred user intent. In some examples,virtual assistant server114 can communicate withexternal services124, such as telephony services, calendar services, information services, messaging services, navigation services, and the like, through network(s)108 for task completion or information acquisition. The I/O interface toexternal services116 can facilitate such communications.

Server system110 can be implemented on one or more standalone data processing devices or a distributed network of computers. In some examples,server system110 can employ various virtual devices and/or services of third-party service providers (e.g., third-party cloud service providers) to provide the underlying computing resources and/or infrastructure resources ofserver system110.

Although the functionality of the virtual assistant is shown inFIG. 1 as including both a client-side portion and a server-side portion, in some examples, the functions of the assistant can be implemented as a standalone application installed on a user device. In addition, the division of functionalities between the client and server portions of the virtual assistant can vary in different examples. For instance, in some examples, the client executed onuser device102 can be a thin-client that provides only user-facing input and output processing functions, and delegates all other functionalities of the virtual assistant to a backend server.

User Device

FIG. 2 is a block diagram of auser device102 according to various examples. As shown,user device102 can include amemory interface202, one ormore processors204, and aperipherals interface206. The various components inuser device102 can be coupled together by one or more communication buses or signal lines.User device102 can further include various sensors, subsystems, and peripheral devices that are coupled to theperipherals interface206. The sensors, subsystems, and peripheral devices gather information and/or facilitate various functionalities ofuser device102.

For example,user device102 can include amotion sensor210, alight sensor212, and aproximity sensor214 coupled to peripherals interface206 to facilitate orientation, light, and proximity sensing functions. One or moreother sensors216, such as a positioning system (e.g., a GPS receiver), a temperature sensor, a biometric sensor, a gyroscope, a compass, an accelerometer, and the like, are also connected toperipherals interface206, to facilitate related functionalities.

In some examples, acamera subsystem220 and anoptical sensor222 can be utilized to facilitate camera functions, such as taking photographs and recording video clips. Communication functions can be facilitated through one or more wired and/orwireless communication subsystems224, which can include various communication ports, radio frequency receivers and transmitters, and/or optical (e.g., infrared) receivers and transmitters. Anaudio subsystem226 can be coupled tospeakers228 and amicrophone230 to facilitate voice-enabled functions, such as voice recognition, voice replication, digital recording, and telephony functions.

In some examples,user device102 can further include an I/O subsystem240 coupled toperipherals interface206. I/O subsystem240 can include atouch screen controller242 and/or other input controller(s)244. Touch-screen controller242 can be coupled to atouch screen246.Touch screen246 and thetouch screen controller242 can, for example, detect contact and movement or break thereof using any of a plurality of touch sensitivity technologies, such as capacitive, resistive, infrared, and surface acoustic wave technologies, proximity sensor arrays, and the like. Other input controller(s)244 can be coupled to other input/control devices248, such as one or more buttons, rocker switches, a thumb-wheel, an infrared port, a USB port, and/or a pointer device such as a stylus.

In some examples,user device102 can further include amemory interface202 coupled tomemory250.Memory250 can include any electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, a portable computer diskette (magnetic), a random access memory (RAM) (magnetic), a read-only memory (ROM) (magnetic), an erasable programmable read-only memory (EPROM) (magnetic), a portable optical disc such as CD, CD-R, CD-RW, DVD, DVD-R, or DVD-RW, or flash memory such as compact flash cards, secured digital cards, USB memory devices, memory sticks, and the like. In some examples, a non-transitory computer-readable storage medium ofmemory250 can be used to store instructions (e.g., for performing some or all ofprocess500, described below) for use by or in connection with an instruction execution system, apparatus, or device, such as a computer-based system, processor-containing system, or other system that can fetch the instructions from the instruction execution system, apparatus, or device, and execute the instructions. In other examples, the instructions (e.g., for performingprocess500, described below) can be stored on a non-transitory computer-readable storage medium ofserver system110, or can be divided between the non-transitory computer-readable storage medium ofmemory250 and the non-transitory computer-readable storage medium ofserver system110. In the context of this document, a “non-transitory computer readable storage medium” can be any medium that can contain or store the program for use by or in connection with the instruction execution system, apparatus, or device.

In some examples, thememory250 can store anoperating system252, acommunication module254, a graphicaluser interface module256, asensor processing module258, aphone module260, andapplications262.Operating system252 can include instructions for handling basic system services and for performing hardware-dependent tasks.Communication module254 can facilitate communicating with one or more additional devices, one or more computers, and/or one or more servers. Graphicaluser interface module256 can facilitate graphic user interface processing.Sensor processing module258 can facilitate sensor-related processing and functions.Phone module260 can facilitate phone-related processes and functions.Application module262 can facilitate various functionalities of user applications, such as electronic messaging, web browsing, media processing, navigation, imaging, and/or other processes and functions.

As described herein,memory250 can also store client-side virtual assistant instructions (e.g., in a virtual assistant client module264) and various user data266 (e.g., user-specific vocabulary data, preference data, and/or other data, such as the user's electronic address book, to-do lists, shopping lists, etc.) to provide the client-side functionalities of the virtual assistant.

In various examples, virtualassistant client module264 can be capable of accepting voice input (e.g., speech input), text input, touch input, and/or gestural input through various user interfaces (e.g., I/O subsystem240,audio subsystem226, or the like) ofuser device102. Virtualassistant client module264 can also be capable of providing output in audio (e.g., speech output), visual, and/or tactile forms. For example, output can be provided as voice, sound, alerts, text messages, menus, graphics, videos, animations, vibrations, and/or combinations of two or more of the above. During operation, virtualassistant client module264 can communicate with the virtual assistant server usingcommunication subsystem224.

In some examples, virtualassistant client module264 can utilize the various sensors, subsystems, and peripheral devices to gather additional information from the surrounding environment ofuser device102 to establish a context associated with a user, the current user interaction, and/or the current user input. In some examples, virtualassistant client module264 can provide the contextual information or a subset thereof with the user input to the virtual assistant server to help infer the user's intent. The virtual assistant can also use the contextual information to determine how to prepare and deliver outputs to the user.

In some examples, the contextual information that accompanies the user input can include sensor information, such as lighting, ambient noise, ambient temperature, images or videos of the surrounding environment, distance to another object, and the like. The contextual information can further include information associated with the physical state of user device102 (e.g., device orientation, device location, device temperature, power level, speed, acceleration, motion patterns, cellular signal strength, etc.) or the software state of user device102 (e.g., running processes, installed programs, past and present network activities, background services, error logs, resources usage, etc.). Any of these types of contextual information can be provided to thevirtual assistant server114 as contextual information associated with a user input.

In some examples, virtualassistant client module264 can selectively provide information (e.g., user data266) stored onuser device102 in response to requests from thevirtual assistant server114. Virtualassistant client module264 can also elicit additional input from the user via a natural language dialogue or other user interfaces upon request byvirtual assistant server114. Virtualassistant client module264 can pass the additional input tovirtual assistant server114 to help virtualassistant server114 in intent inference and/or fulfillment of the user's intent expressed in the user request.

In various examples,memory250 can include additional instructions or fewer instructions. Furthermore, various functions ofuser device102 can be implemented in hardware and/or in firmware, including in one or more signal processing and/or application-specific integrated circuits.

Output Dialog Encoding

As mentioned above, system100 can be used to implement a virtual assistant capable of interpreting natural language input in spoken and/or textual form to infer user intent, and performing actions based on the inferred user intent. In some examples, the performed actions can include generating and outputting an output dialog in text and/or audio form. For example, in response to the query “What is the temperature in New York?”, a virtual assistant implemented on a user device can display the text “The temperature in New York is 72° F.” In order for the virtual assistant to produce this output dialog in natural language form, instructions that define the contents and format of the dialog can be provided to the virtual assistant.FIG. 3 shows a visual representation of adialog generation file300 that can be used to provide a virtual assistant with such information.

Generally,dialog generation file300 can be used to generate an output dialog (e.g., “Good morning, John.”) by defining multiple output segments that represent a portion of the output dialog (e.g., a first output segment “Good morning,” and a second output segment “John.”).Dialog generation file300 can further define the possible outputs that can be used in each output segment (e.g., “Good morning,” “Good afternoon,” or “Good evening,” for the first output segment), how to select one of the possible output segments, and how to format the output segments to generate the output dialog. As discussed in greater detail below,dialog generation file300 can define these parameters for generating the output dialog using various types of functional blocks302-310 arranged hierarchically and/or in a particular order to generate a desired output dialog.

As shown inFIG. 3,dialog generation file300 can include multiple types of functional blocks302-310 that can output or return an output segment of text. An output segment of text can include a string of one or more characters or words and can be ordered amongst the other output segments of text based on an ordering of the functional blocks withindialog generation file300. In some examples, the output dialog can be generated by concatenating the output segments of text based on the order of the output segments. For example, ifdialog generation file300 includesrandom block305 outputting the output segment of text “Hi,” followed bytext block302 outputting the output segment of text “how are you doing?”, the output dialog generated fromdialog generation file300 can include the text “Hi, how are you doing?”.

Dialog generation file300 can include any number of and type of functional blocks. In some examples, the functional blocks can be arranged in a hierarchy such that one functional block (e.g., the child functional block) is included within another functional block (e.g., the parent functional block). In these examples, the output segment of text output by the child functional block can be returned as an input to the parent functional block and used by the parent functional block in a manner that depends on a type of the parent functional block (e.g., a type corresponding to blocks302-310, described below). In some examples, a functional block can include any number of other functional blocks. Additionally, the hierarchy formed by the nested functional blocks can include any number of levels. For example, a first functional block can be included within a second functional block, which can be included within a third functional block, and so on.

One type of functional block that can be included indialog generation file300 istext block302.Text block302 can be configured to include desired text represented as a predetermined string of characters or words, such as “Good morning,” and/or a variable representing a number or a string of characters or words, such as <name> having a value of “John.” The formatting of the predetermined string of characters or words with a variable (e.g., “Good morning,”<name>) can be referred to as a “template.” When executed,text block302 can be configured to output the included text as an output segment of text. For example, atext block302 that includes the text “Okay, here are your search results:” can output that text as an output segment of text when executed.

Dialog generation file300 can further includefirst block303, which can be configured to include an ordered list of one or more options (e.g.,Option 1,Option 2, and Option 3). Each option can include one or more of the functional blocks shown in FIG.3. When executed,first block303 can select the first option (Option 1) from the ordered list and can call the functional block associated with that option. First block303 can then output the text received from the called functional block (e.g., the output segment produced by the called functional block) as an output segment of text. If the selection of the first option is successful (e.g., successfully calls another functional block), the operation offirst block303 can end. However, if the selection of the first option fails,first block303 can then attempt to select the next option from the ordered list (Option 2). This process can repeat for all options in the ordered list until an option is successfully selected. In the event that no option is successfully selected,first block303 can return an error. Whilefirst block303 is shown having three options, it should be appreciated thatfirst block303 can include any number of options.

To illustrate the operation offirst block303,Option 1 offirst block303 can include atext block302 that includes the text “Hello,”Option 2 offirst block303 can include atext block302 that includes the text “Hi,” andOption 3 offirst block303 can include switch block304 (described below). Upon execution,first block303 can first attempt to call thefirst text block302 and output the text “Hello” returned by thefirst text block302 as an output segment of text. If successful, the operation ofblock303 can end. However, if unable to output the text “Hello” as an output segment of text for any reason,first block303 can then attempt to call thesecond text block302 and output the text “Hi” returned by thesecond text block302 as an output segment of text. If successful, the operation ofblock303 can end. However, if unable to output the text “Hi” as an output segment of text for any reason,first block303 can then attempt to callswitch block304. Ifswitch block304 is successfully executed and returns a text output,first block303 can output the text returned fromswitch block304 as an output segment of text and the operation ofblock303 can end. However, if unable to successfully callswitch block304 for any reason,first block303 can return an error because it reached the end of its ordered list without successfully selecting one of the three options.

Dialog generation file300 can further includeswitch block304, which can be configured to include an input parameter (e.g., <parameter>) and a list of one or more choices (e.g.,Option 1,Option 2, and Option 3) that are associated with potential values of the input parameter (e.g., <A> and <B>) or a default condition (e.g., Default). Each option can include one of the functional blocks shown inFIG. 3. When executed, switch block304 can select one of the options based on the value of the input parameter matching an associated value of one of the options and can call the functional block associated with that option.Switch block304 can then output the text received from the called functional block (e.g., the output segment produced by the called functional block) as an output segment of text. For example, switch block304 can selectOption 1 if the value of <parameter> is equal to the value of <A>, can selectOption 2 if the value of <parameter> is equal to the value of <B>, and can selectOption 3 if the value of <parameter> is not equal to the values of <A> or <B>. Whileswitch block304 is shown as having three options, it should be appreciated that switch block304 can include any number of options.

In some examples, <parameter> can include any desired variable representing information that is to be presented to the user (e.g., search query results, etc.) or contextual information, such as sensor information, information associated with the physical state of a user device, information associated with a software state of the user device, or the like. For example, the contextual information can include a type of user device used to operate the virtual assistant, an operational state of the user device, a current time, a location of the user device, an orientation of the user device, or an orientation of a user with respect to the user device.

To illustrate the operation ofswitch block304, <parameter> ofswitch block304 can represent a number of results generated from a search query, <A> can represent the value 0,Option 1 ofswitch block304 can include atext block302 that includes the text “Your search produced no results,”<B> can represent thevalue 1,Option 2 ofswitch block304 can include atext block302 that includes the text “Your search produced 1 result,” andOption 3 ofswitch block304 can include atext block302 that includes the text “Your search produced multiple results.” In this example, switch block304 can call thefirst text block302 and output the text “Your search produced no results” returned by thefirst text block302 as an output segment of text if the number of results generated from the search query is 0 (<parameter>==0), can call thesecond text block302 and output the text “Your search produced 1 result” returned by thesecond text block302 as an output segment of text if the number of results generated from the search query is 1 (<parameter>==1), and can call thethird text block302 and output the text “Your search produced no results” returned by thethird text block302 as an output segment of text if the number of results generated from the search query is not equal to 0 or 1 (<parameter>!=0 && <parameter>!=1).

Dialog generation file300 can further includerandom block305, which can be configured to include a list of one or more options (e.g.,Option 1,Option 2, and Option 3). Each option can include one of the functional blocks shown inFIG. 3. When executed,random block305 can randomly select one of the options from its list and can call the functional block associated with that option. Random block305 can then output text received from the called functional block (e.g., the output segment produced by the called functional block) as an output segment of text. If the selection of the randomly selected block is successful (e.g., successfully calls another functional block), the operation ofrandom block305 can end. However, if the selection of the randomly selected block fails,random block305 can randomly select another option from the list that was not previously selected. This process can repeat until an option is successfully selected. In the event that no option is successfully selected and there are no options remaining in the list that were not previously selected,random block305 can return an error. Whilerandom block305 is shown having three options, it should be appreciated thatrandom block305 can include any number of options.

In some examples, the options ofrandom block305 can be assigned a weighting factor that can be used byrandom block305 to increase or decrease the probability that a particular option is selected. In the absence of assigned weighting factors, the probability of selecting each option can be the same (e.g., equal weighting factors for all options).

To illustrate the operation ofrandom block305,Option 1 ofrandom block305 can include atext block302 that includes the text “Hello,”Option 2 ofrandom block305 can include atext block302 that includes the text “Hi,” andOption 3 ofrandom block305 can includeswitch block304. Upon execution,random block305 can randomly select one of the options (e.g., Option 2) and attempt to call thesecond text block302 and output the text “Hi” returned by thesecond text block302 as an output segment of text. If successful, the operation ofrandom block305 can end. However, if unable to output the text “Hi” as an output segment of text for any reason,random block305 can then randomly select another option from the list that was not previously selected (e.g., select fromOption 1 and Option 3). If the random selection results in a selection ofOption 1,random block305 can attempt to call thefirst text block302 and output the text “Hello” returned by thefirst text block302 as an output segment of text. If successful, the operation ofrandom block305 can end. However, if unable to output the text “Hello” as an output segment of text for any reason,random block305 can then randomly select another option from the list that was not previously selected (e.g., select from Option 3). SinceOption 3 is the only remaining option that was not previously selected,random block305 can selectOption 3 and attempt to callswitch block304. Ifswitch block304 is successfully executed and returns a text output,random block305 can output the text returned fromswitch block304 as an output segment of text and the operation ofrandom block305 can end. However, if unable to successfully callswitch block304 for any reason,random block305 can return an error because no options were successfully selected and no options remain in the list that were not previously selected.

Dialog generation file300 can further includeopt block306, which is an optional block that can be used to ignore any failures returned by one of its included functional blocks, such asfirst block303 andrandom block305. Opt block306 can include one or more of the functional blocks shown inFIG. 3. When executed, optblock306 can call the functional blocks contained therein and can output text received from the called functional blocks as an output segment of text. If a functional block included withinopt block306 returns an error (e.g.,first block303 orrandom block305 runs out of options), optblock306 can ignore that error to prevent the error from stopping the generation of the output dialog.

To illustrate the operation ofopt block306, optblock306 can includefirst block303, which can includeOption 1 having atext block302 that includes the text “Hello,”Option 2 having atext block302 that includes the text “Hi,” andOption 3 having atext block302 that includes the text “Greetings.” Upon execution, optblock306 can callfirst block303, which can first attempt to call thefirst text block302 and output the text “Hello” returned by thefirst text block302 as an output segment of text. If successful,first block303 can return the text “Hello” to optblock306, which can then output the text “Hello” as an output segment of text. However, if unable to output the text “Hello” as an output segment of text for any reason,first block303 can then attempt to call thesecond text block302 and output the text “Hi” returned by thesecond text block302 as an output segment of text. If successful,first block303 can return the text “Hi” to optblock306, which can then output the text “Hi” as an output segment of text. However, if unable to output the text “Hi” as an output segment of text for any reason,first block303 can then attempt to call thethird text block302 and output the text “Greetings” returned by thethird text block302 as an output segment of text. If successful,first block303 can return the text “Greetings” to optblock306, which can then output the text “Greetings” as an output segment of text. However, if unable to output the text “Greetings” as an output segment of text for any reason,first block303 can return an error to opt block306 because it reached the end of its ordered list without successfully selecting one of the three options. Opt block306 can ignore the error and the operation ofopt block306 can end without outputting or returning an output segment of text.

Dialog generation file300 can further includeconditional block307, which is a functional block that can include one or more of the functional blocks shown inFIG. 3 and can be executed only when a conditional statement associated withconditional block307 is satisfied. For example,conditional block307 can be associated with the conditional if-statement (if A==B). In this example,conditional block307 can only be executed if A is equal to B. If A is not equal to B, conditional block307 (and any functional blocks included within conditional block307) may not be executed. When executed,conditional block307 can call the functional blocks contained therein and can output text received from the called functional blocks as an output segment of text.

In some examples, the conditional statement associated withconditional block307 can be based on information that is to be presented to the user (e.g., search query results, etc.) or contextual information, such as sensor information, information associated with the physical state of a user device, information associated with a software state of the user device, or the like. For example, the contextual information can include a type of user device used to operate the virtual assistant, an operational state of the user device, a current time, a location of the user device, an orientation of the user device, or an orientation of a user with respect to the user device.

To illustrate the operation ofconditional block307,conditional block307 can be associated with the if-statement (if name==available) and can include atext block302 that includes the text <name>. In this example, if the user's name is available, thenconditional block307 can be executed and can calltext block302 and output the value of <name> returned bytext block302 as an output segment of text. Alternatively, if the user's name is not available, thenconditional block307 may not be executed and no output segment of text may be generated byconditional block307.

Typically, any output segment of text generated bytext block302,first block303,switch block304,random block305, optblock306, andconditional block307 can be identified as being both a spoken output and a printed output (e.g., the text can be presented to the user in both audio and visual format by a user device). To cause the string of words or characters in an output segment to be only printed (e.g., only displayed to the user),dialog generation file300 can include print only block308. Print only block308 can include one or more of the functional blocks shown inFIG. 3. When executed, print only block308 can call the functional blocks contained therein and can output text returned from the called functional blocks as an output segment of text. However, unlike the other functional blocks ofFIG. 3, print only block308 can identify its output segment of text as being text that should only be printed (e.g., displayed) when presented to the user.

To illustrate the operation of print only block308, print only block308 can include atext block302 that includes the text “Good afternoon.” In this example, print only block308 can calltext block302 and output the text “Good afternoon” returned bytext block302 as an output segment of text that is identified as being only a printed output.

Similarly, to cause the string of words or characters in an output segment to be only spoken (e.g., only audibly read to the user),dialog generation file300 can include speak only block309. Speak only block309 can include one or more of the functional blocks shown inFIG. 3. When executed, speak only block309 can call the functional blocks contained therein and can output text returned from the called functional blocks as an output segment of text. However, unlike the other functional blocks ofFIG. 3, speak only block309 can identify its output segment of text as being text that should only be spoken (e.g., audibly presented) when presented to the user.

To illustrate the operation of speak only block309, speak only block300 can include atext block302 that includes the text “Good afternoon.” In this example, speak only block309 can calltext block302 and output the text “Good afternoon” returned bytext block302 as an output segment of text that is identified as being only a spoken output.

Dialog generation file300 can further includebreak block310, which represents a break function that indicates a separation in the formatting of the output dialog. For example, if the output dialog is intended to be displayed within a speech bubble, the break function ofbreak block310 can indicate that a new speech bubble should be created for the output segments of text following the break.

Since adialog generation file300 can be used to generate an output dialog for conveying a particular type of information or message to a user, multiple different dialog generation files300 can be created to generate output dialogs that are appropriate for a variety of situations. Additionally, sincedialog generation file300 can be language-specific, a differentdialog generation file300 can be created for each language in which an output dialog is to be created.

Whiledialog generation file300 is shown and described above as having functional blocks performing various functions, it should be appreciated thatdialog generation file300 can be implemented using any desired programming language. In some examples, a markup-based language, such as XML, can be used.

FIG. 4 illustrates a visual representation of an exampledialog generation file400 that can be used to generateoutput dialog432 using the functional blocks302-310 shown inFIG. 3. As shown inFIG. 4,dialog generation file400 includes hierarchically structured functional blocks401-414. The functional blocks are arranged in order from left to right, and functional blocks contained within other functional blocks represent functional blocks that are child functional blocks to the enclosing parent functional blocks. Additionally, functional blocks401-414 having names matching those of the functional blocks shown inFIG. 3 can represent instances of those functional blocks.

When a call is made todialog generation file400 to createoutput dialog432, the first block in the highest level of the hierarchy (opt block401) can be executed. When executed, optblock401 can callrandom block402 contained therein. Random block402 can then randomly select one oftext block403,404, and405 based on their associated weighting factors (e.g., weighting factors50,30, and20). For example,text block403 can have a 50% chance of being selected,text block404 can have a 30% chance of being selected, and text block405 can have a 20% chance of being selected. Oncerandom block402 randomly selects one of the text blocks, a call can be made to the selected text block. For example, iftext block405 is selected,text block405 can be called. In response,text block405 can return the text “Hi” torandom block402. Random block402 can then return the text “Hi” to optblock401. Opt block401 can then output the text “Hi” as an output segment of text. Sinceopt block401 is a functional block at the highest level of the functional block hierarchy, the output segment of text generated byopt block401 can represent anoutput segment430 ofdialog generation file400.

In some examples, the weighted random function performed byrandom block402 can consider weights (some of which can be defaults) of all blocks within the random block to determine the random selection. For example, ifrandom block402 includes threeelements having weights 2, 5, and 42, then the elements can have a probability of being selected of 2/49, 5/49, and 42/49, respectively.

Upon completion ofopt block401, execution can proceed to the next functional block in the highest level of the hierarchy (print only block406). When executed, print only block406 can callfirst block407 contained therein. First block407 can then select and call the functional block associated with its first option (conditional block408). If successful,first block407 can return the text received fromconditional block408 to print only block406. However, if the selection ofconditional block408 fails (e.g., if the variable <user> is not available), thefirst block407 can attempt to calltext block410 and output the text received fromtext block410 to print only block407. For example, if the variable <name> (representing the value “John”) is available,conditional block408 can calltext block409.Text block409 can return the text “, John.” toconditional block408, which can then return the text “, John.” to print only block406. Print only block406 can then output the text “, John.” as an output segment of text that is identified as being a print only output. Since print only block406 is a functional block at the highest level of the functional block hierarchy, the output segment of text generated by print only block406 can represent anoutput segment430 ofdialog generation file400.

Upon completion ofprint block406, execution can proceed to the next functional block in the highest level of the hierarchy (speak only block411). When executed, speak only block411 can callswitch block412 contained therein.Switch block412 can then select and call the functional block associated with one of its two options based on the value of the variable <AM/PM>.Switch block412 can then return the text received from the called functional block to speak only block411. For example, if the variable <AM/PM> represents the value “AM,”switch block412 can calltext block413.Text block413 can return the text “How are you doing this morning” to switchblock412, which can then return the text “How are you doing this morning” to speak only block411. Speak only block411 can then output the text “How are you doing this morning” as an output segment of text that is identified as being a speak only output. Since speak only block411 is a functional block at the highest level of the functional block hierarchy, the output segment of text generated by speak only block411 can represent anoutput segment430 ofdialog generation file400.

Upon completion of all of the functional blocks at the highest level of the hierarchy, anoutput dialog432 can be generated based on theoutput segments430 of thedialog generation file400. Theoutput segments430 can have an associated ordering based on the order of the functional blocks (that generated them) withindialog generation file400. For example, the output segment containing “Hi” output byopt block401 can be first, the output segment containing “, John.” output by print only block406 can be second, and the output segment containing “How are you doing this morning?” output by speak only block411 can be third.Output dialog432 can be generated by concatenatingoutput segments430 in order from the first output segment to the last. Thus,output dialog432 can include “Hi, John. How are you doing this morning?”. When presented to the user, the words “Hi, John” can be displayed to the user. The words “Hi. How are you doing this morning?” can be spoken to the user.

As illustrated by the example shown inFIG. 4, theoutput segments430 used to createoutput dialog432 can be created independent of each other. For example, the generation of one output segment may not depend on the generation of another segment. In other examples, oneoutput segment430 can be generated using a conditional statement that depends on the generation of anotheroutput segment430.

Output Dialog Generation

FIG. 5 illustrates anexemplary process500 that can be performed to generate an output dialog. In some examples,process500 can be performed using a system similar or identical to system100, shown inFIG. 1.

Atblock502, an audio input including user speech can be received at a user device. In some examples, a user device (e.g., user device102) can receive audio input that includes a user's speech via a microphone (e.g., microphone230). The microphone can convert the audio input into an analog or digital representation, and provide the audio data to one or more processors (e.g., processor(s)204). In some examples, the user device can transmit data representing the audio input to one or more servers, such asserver system110.

Atblock504, the user speech of the audio input can be converted into a textual representation of the user speech. The user speech can be converted using any known speech-to-text conversion process. In some examples, the user speech can be converted into the textual representation locally on theuser device102. In other examples where the user device transmits data representing the audio input to one or more servers for processing, the one or more servers can receive the data representing the audio input and can instead perform the speech-to-text conversion process.

Atblock506, the textual representation of user speech generated atblock504 can be received or accessed, and a response to the textual representation can be determined. The response can include information that is to be provided to the user in response to the textual representation of user speech. For example, a response to a search query can include information relating to the search query results (e.g., number of results, the actual results, etc.). The response can further include an identification of a dialog generation file that is to be used to format the output dialog to convey some or all of the information of the response to the user.

In some examples, determining the response to the textual representation can include determining a user intent based on the textual representation of user speech. As discussed in greater detail in Applicants' U.S. Utility application Ser. No. 12/987,982 for “Intelligent Automated Assistant,” filed Jan. 10, 2011, determining user intent can include analyzing, by processingmodules118, the textual representation of user speech to identify a set of syntactic or semantic parse results. The syntactic parse results can include parse results that associate data in the user input with structures that represent syntactic parts of speech, clauses, and phrases including multiword names, sentence structure, and/or other grammatical graph structures. The semantic parse results can include parse results that associate data in the user input with structures that represent concepts, relationships, properties, entities, quantities, propositions, and/or other representations of meaning and user intent. Determining user intent can further include disambiguating among alternative syntactic or semantic parse results based on the contextual representation of user speech, other user input, and/or contextual information.

Determining the response to the textual representation of user speech can further include identifying a task flow based on the determined user intent and executing the identified task flow. The task flow can include a set of steps and associated parameters to accomplish the determined user intent. In some examples, the task flow can include the step of selecting an appropriate dialog generation file for outputting the results of the task flow to the user. The dialog generation file can be selected based on the type of information that is to be provided to the user, the user's native language, other contextual information, and the like. The steps of the task flow can be executed by invoking programs, methods, services, APIs, or the like.

Atblock508, a plurality of output segments can be generated based on the response determined atblock506. The output segments can include a string of one or more characters or words that can be used to generate the output dialog presented to the user. In some examples, one or more of the output segments can be generated by selecting one output from multiple possible outputs. Each output can include a string of one or more characters or words and/or a variable. For example, a first possible output for an output segment can include “Hello,” a second possible output for the output segment can include “Hi,” and a third possible output for the output segment can include “Hi,”<user>, where <user> is a variable representing the user's name. Thus, generating the output segment can include selecting one of the three possible outputs and using the selected output as the output segment.

In some examples, the dialog generation file selected atblock506 can be used to generate the plurality of output segments. The type, content, and ordering of the functional blocks within the dialog generation file can define the possible outputs for each output segment, an ordering of the output segments within the output dialog, and format for the output dialog to be displayed or otherwise presented to a user and it can be used to generate the output dialog. As discussed above with respect toFIGS. 3-4, the functional blocks at the highest level of the hierarchy in the dialog generation file (and their child functional blocks) can be executed based on their respective ordering in the file to generate the plurality of output segments.

In some examples, an output segment can be generated based on a random block of the dialog generation file to randomly select an output from multiple outputs as discussed above with respect toFIGS. 3-4. The multiple outputs can be uniformly or non-uniformly weighted using weighting factors associated with the outputs. In other examples, an output segment can also be generated based on a first block of the dialog generation file to select an output from multiple outputs based on a predetermined order of the multiple outputs as discussed above with respect toFIGS. 3-4. In yet other examples, an output segment can be generated based on a switch block of the dialog generation file to select an output from multiple outputs based on a value of an input parameter as discussed above with respect toFIGS. 3-4.

In some examples, some or all of the possible outputs for an output segment can be associated with a conditional statement of a conditional block and the output can be selected from the plurality of possible outputs based on whether or not the condition is satisfied as discussed above with respect toFIGS. 3-4. The conditional statement can depend on the information to be provided to the user determined atblock506 and/or contextual information, such as sensor information, information associated with the physical state of a user device, information associated with a software state of the user device, or the like. For example, the contextual information can include a type of user device used to operate the virtual assistant, an operational state of the user device, a current time, a location of the user device, an orientation of the user device, or an orientation of a user with respect to the user device.

In some examples, the functional blocks of the dialog generation file can further define a format for outputting the plurality of output segments to an output dialog. For example, an output segment generated from a print only block can be tagged as being an output segment that should only be displayed to a user. An output segment generated from a speak only block can be tagged as being an output segment that should only be audibly spoken to a user. Additionally, a break block can indicate that an audio and/or visual break should be included between the output segment generated from the previous functional block and the output segment generated from the subsequent output block.

In some examples, the output segments can be created independent of each other. For example, the generation of one output segment may not depend on the generation of another segment. In other examples, one output segment can be generated using a conditional statement that depends on the generation of another output segment, or a condition can be placed on a functional block included within a random block.

Atblock510, an output dialog can be generated based on the plurality of segments. In some examples, generating the output dialog can include concatenating the plurality of output segments based on an order of the output segments. For example, if a first output segment includes “Hi,” a second output segment includes “,”<user>, and a third output segment includes “. How are you doing today?”, generating the output dialog based on these three output segments can include concatenating them based on their order (e.g., first, second, and third) to produce the output dialog “Hi, John. How are you doing today?”.

In some examples, as mentioned above, the output segments can be identified as being only a spoken output, only a printed output, or both a spoken and printed output. In these examples, generating the output dialog can include generating a spoken output dialog by concatenating the output segments identified as being only a spoken output, generating a printed output dialog by concatenating the output segments identified as being only a printed output, and generating a spoken and printed output dialog by concatenating the output segments identified as being both a printed and spoken output. Alternatively, a single output dialog can be generated and the words corresponding to the spoken only outputs can be tagged or otherwise identified as being only spoken, the words corresponding to the printed only outputs can be tagged or otherwise identified as being only printed, and the words corresponding to the printed and spoken outputs can be tagged or otherwise identified as being both spoken and printed.

In some examples, as mentioned above,process500 can include tagging one or more words of the textual user input using synonyms or can include mapping a word, or a synonym for that word, to a single semantic value, and from that value to a grammatically appropriate synonym for textual output. In these examples, generating the output dialog atblock510 can include replacing an occurrence of a tagged word with the synonym or alternate word of the tag. For example, if the textual representation of user speech received atblock502 included “Forward the message to John”, the word “message” may have been tagged with the word “email.” If the plurality of output segments generated atblock508 included a first output segment “Your message has been forwarded to” and a second output segment “John”, block510 can include concatenating the first and second segments to form the output dialog, “Your message has been forwarded to John.” Block510 can further include replacing the word “message” with the word “email” based on the tag to form the output dialog, “Your email has been forwarded to John.” Alternatively, the word “message” can be replaced with the word “email” in the output segment prior to concatenating the segments.

In some examples,process500 can further include transmitting the output dialog generated atblock510 to a user device. The user device can receive the output dialog and present it to the user. In some examples, the output dialog can be displayed to the user on a display of the user device and/or presented to the user in audio form. For example, as mentioned above, the output segments can be identified as being only a spoken output, only a printed output, or both a spoken and printed output. In some examples, the output dialog can be identified as being a spoken output dialog, a printed output dialog, and a spoken and printed output dialog. The spoken output dialog can be presented to the user in audio form, but not displayed. The printed output dialog can be displayed to the user, but not presented to the user in audio form. The spoken and printed output dialog can be presented to the user in audio form and displayed to the user. In other examples, a single output dialog having words tagged or otherwise identified as being only spoken, only printed, or both spoken and printed can be received. In these examples, the words tagged or otherwise identified as being only spoken can be presented to the user in audio form, but not displayed. The words tagged or identified as being only printed can be displayed to the user, but not presented in audio form. The words tagged or identified as being both spoken and printed can be presented to the user in audio form and displayed to the user.

In some examples, as mentioned above, the output dialog can include formatting information that details how the output dialog should be presented to the user. For example, an output dialog can include breaks that represent separations in the output dialog that should be displayed or otherwise conveyed to the user. For example, a break in an output dialog can indicate that a new speech bubble or other visual indication of a break in the output dialog should be created for the text following the break.FIG. 6 illustrates anexemplary conversation interface600 between a user and a virtual assistant. As shown, a textual representation of a user's spoken input is represented byspeech bubble602. The virtual assistant's output dialog response is represented byspeech bubbles604 and606. In this example, the output dialog provided to the user device can include “Hi, John.”<break>“How are you doing this morning?”. The <break> in the output dialog can be generated from a break block to indicate to the user device that anew speech bubble606 should be created to display the text following the <break>. The conversation shown inconversation interface600 can be generated fromdialog generation file400 if a break block is inserted between print only block406 and speak only block411.

Usingprocess500, an output dialog can be generated by forming and combining multiple output segments. Each output segment can be configured to be populated with one of multiple strings of characters or words to provide an output that can be dynamically generated based on the information being provided to the user and/or contextual information. Generating output dialogs in this way also obviates the need to program a virtual assistant with every candidate sentence that is to be output by the virtual assistant.

It should be appreciated that the blocks ofprocess500 can be performed onuser device102,server system110, or a combination ofuser device102 andserver system110. For instance, in some examples, all blocks ofprocess500 can be performed onuser device102. In other examples, all blocks ofprocess500 can be performed atserver system110. In yet other examples, some blocks ofprocess500 can be performed atuser device102, while other blocks ofprocess500 can be performed atserver system110.

Electronic Device

In accordance with some examples,FIG. 7 shows a functional block diagram of anelectronic device700 configured in accordance with the principles of the various described examples. The functional blocks of the device can be implemented by hardware, software, or a combination of hardware and software to carry out the principles of the various described examples. It is understood by persons of skill in the art that the functional blocks described inFIG. 7 can be combined or separated into sub-blocks to implement the principles of the various described examples. Therefore, the description herein optionally supports any possible combination or separation or further definition of the functional blocks described herein.

As shown inFIG. 7,electronic device700 can include a touchscreen display unit702 configured to display a user interface and to receive touch input, and asound receiving unit704 configured to receive sound input. In some examples,electronic device700 can include aspeaker unit706 configured to generate sound.Electronic device700 can further include aprocessing unit708 coupled to touchscreen display unit702 and sound receiving unit704 (and, optionally, coupled to speaker unit706). In some examples, processingunit708 can include atext accessing unit710, aresponse determination unit712, an outputsegment generation unit714, an output dialog generation unit716, atagging unit718, an audiodata receiving unit720, a speech totext conversion unit722, and a transmittingunit724.

Processing unit708 can be configured to access a textual representation of user speech (e.g., using text accessing unit710).Processing unit708 can be further configured to determine (e.g., using response determination unit712) a response to the textual representation of user speech.Processing unit708 can be further configured to generate (e.g., output segment generation unit714) a plurality of output segments based on the determined response to the textual representation of user speech, wherein each of the plurality of output segments comprises one or more words.Processing unit708 can be further configured to generate (e.g., output dialog generation unit716) an output dialog based on the plurality of output segments and an order associated with the plurality of output segments.

In some examples, processingunit708 can be configured to determine the response to the textual representation of user speech by determining (e.g., using response determination unit712) a user intent based on the textual representation of user speech; identifying a task flow based on the determined user intent; and executing the identified task flow to determine the response to the textual representation of user speech.

In some examples, processingunit708 can be configured to generate the plurality of output segments (e.g., using output segment generation unit714) based on the determined response to the textual representation of user speech by selecting an output from a plurality of possible outputs for an output segment of the plurality of output segments, the selected output comprising the one or more words of the output segment.

In some examples, processingunit708 can be configured to select (e.g., using output segment generation unit714) the output randomly from the plurality of possible outputs.

In some examples, each of the plurality of possible outputs can be associated with a weighting factor, and whereinprocessing unit708 can be configured to select (e.g., using output segment generation unit714) the output randomly from the plurality of possible outputs based on the weighting factors.

In some examples, processingunit708 can be configured to select (e.g., using output segment generation unit714) the output from the plurality of possible outputs based on a predetermined order of the plurality of possible outputs.

In some examples, the plurality of possible outputs are associated with one or more conditions, and whereinprocessing unit708 can be configured to select (e.g., using output segment generation unit714) the output from the plurality of possible outputs further based on whether or not the one or more conditions are satisfied.

In some examples, the response to the textual representation of user speech can include information to be provided in response to the textual representation of user speech, and wherein the one or more conditions depend on the information to be provided in response to the textual representation of user speech. The one or more conditions can depend on contextual information. In some examples, the contextual information can include a type of user device used to operate the virtual assistant, an operational state of the user device, a current time, a location of the user device, an orientation of the user device, or an orientation of a user with respect to the user device.

In some examples, the response to the textual representation of user speech can include an identification of a dialog generation file, and wherein the dialog generation file defines: the order of the plurality of output segments; and possible outputs for each of the plurality of output segments.

In some examples, each of the possible outputs can be defined using a template, and wherein each template comprises one or more of a string of words and a variable word.

In some examples, the dialog generation file can further define a format for displaying the output dialog.

In some examples, each output segment of the plurality of output segments can be identified as being a spoken output segment or a printed output segment, and whereinprocessing unit708 can be configured to generate the output dialog (e.g., using output dialog generation unit716) by generating a spoken output dialog based on output segments identified as being a spoken output segment; and generating a printed output dialog based on output segments identified as being printed output segments.

In some examples, processingunit708 can be configured to generate (e.g., using output segment generation unit714) each output segment of the plurality of output independent of the other output segments.

In some examples, processingunit708 can be configured to generate (e.g., using output dialog generation unit716) the output dialog based on the plurality of output segments by concatenating the plurality of output segments.

In some examples, processingunit708 can be configured to tag (e.g., using tagging unit718), after accessing the textual representation of user speech and before determining the response to the textual representation of user speech, a word of the user speech with a dialog tag.

In some examples, the dialog tag can include a synonym of the tagged word.

In some examples, processingunit708 can be configured to generate (e.g., using output dialog generation unit716) the output dialog based on the plurality of output segments by replacing an occurrence of the tagged word in one or more of the plurality of output segments with the synonym of the tagged word.

In some examples, processingunit708 can be configured to receive (e.g., using audio data receiving unit720), before accessing the textual representation of user speech, data corresponding to an audio input comprising user speech from a user device; perform (e.g., using speech to text conversion unit722), before accessing the textual representation of user speech, speech to text conversion on the data corresponding to the audio input to generate the textual representation of user speech; and transmit (e.g., using transmitting unit724), after generating the output dialog, the output dialog to the user device.

As described above, one aspect of the present technology is the gathering and use of data available from various sources to improve the delivery to users of invitational content or any other content that may be of interest to them. The present disclosure contemplates that in some instances, this gathered data can include personal information data that uniquely identifies or can be used to contact or locate a specific person. Such personal information data can include demographic data, location-based data, telephone numbers, email addresses, home addresses, or any other identifying information.

The present disclosure recognizes that the use of such personal information data, in the present technology, can be used to the benefit of users. For example, the personal information data can be used to deliver targeted content that is of greater interest to the user. Accordingly, use of such personal information data enables calculated control of the delivered content. Further, other uses for personal information data that benefit the user are also contemplated by the present disclosure.

The present disclosure further contemplates that the entities responsible for the collection, analysis, disclosure, transfer, storage, or other use of such personal information data will comply with well-established privacy policies and/or privacy practices. In particular, such entities should implement and consistently use privacy policies and practices that are generally recognized as meeting or exceeding industry or governmental requirements for maintaining personal information data private and secure. For example, personal information from users should be collected for legitimate and reasonable uses of the entity and not shared or sold outside of those legitimate uses. Further, such collection should occur only after receiving the informed consent of the users. Additionally, such entities would take any needed steps for safeguarding and securing access to such personal information data and ensuring that others with access to the personal information data adhere to their privacy policies and procedures. Further, such entities can subject themselves to evaluation by third parties to certify their adherence to widely accepted privacy policies and practices.

Despite the foregoing, the present disclosure also contemplates examples in which users selectively block the use of, or access to, personal information data. That is, the present disclosure contemplates that hardware and/or software elements can be provided to prevent or block access to such personal information data. For example, in the case of advertisement delivery services, the present technology can be configured to allow users to select to “opt in” or “opt out” of participation in the collection of personal information data during registration for services. In another example, users can select not to provide location information for targeted content delivery services. In yet another example, users can select to not provide precise location information, but permit the transfer of location zone information.

Therefore, although the present disclosure broadly covers use of personal information data to implement one or more various disclosed examples, the present disclosure also contemplates that the various examples can also be implemented without the need for accessing such personal information data. That is, the various examples of the present technology are not rendered inoperable due to the lack of all or a portion of such personal information data. For example, content can be selected and delivered to users by inferring preferences based on non-personal information data or a bare minimum amount of personal information, such as the content being requested by the device associated with a user, other non-personal information available to the content delivery services, or publicly available information.

Although examples have been fully described with reference to the accompanying drawings, it is to be noted that various changes and modifications will become apparent to those skilled in the art. Such changes and modifications are to be understood as being included within the scope of the various examples as defined by the appended claims.

Claims (36)

What is claimed is:

1. A method for operating a virtual assistant, performed at an electronic device having one or more processors and a memory storing one or more programs for execution by the one or more processors, the method comprising:

receiving a user speech;

accessing a textual representation of the user speech;

determining an intent corresponding to the textual representation of the user speech;

generating an output dialog responsive to the user speech, comprising:

selecting, in accordance with the determined intent, a dialog template from a plurality of dialog templates, wherein the dialog template includes a plurality of sets of functional blocks, wherein a first set of functional blocks of the plurality of sets of functional blocks defines one or more conditions;

generating a plurality of output segments from the plurality of sets of functional blocks, including:

determining whether the one or more conditions are satisfied; and

in accordance with a determination that the one or more conditions are satisfied, generating, with the first set of functional blocks, a first string for a first output segment of the plurality of output segments; and

concatenating the plurality of output segments according to an order defined by the dialog template; and

outputting the output dialog comprising the concatenated plurality of output segments.

2. The method ofclaim 1, wherein generating an output dialog responsive to the user speech further comprises:

identifying a task flow based on the determined user intent; and

executing the identified task flow to generate the output dialog.

3. The method ofclaim 1, wherein generating the plurality of output segments from the plurality of sets of functional blocks further comprises:

selecting an output from a plurality of possible outputs for a second set of functional blocks of the plurality of sets of functional blocks, wherein an output segment of the plurality of output segments includes one or more words of the selected output.

4. The method ofclaim 3, wherein the output is selected randomly from the plurality of possible outputs.

5. The method ofclaim 4, wherein each of the plurality of possible outputs is associated with a weighting factor, and wherein the output is selected randomly from the plurality of possible outputs based on the weighting factors.

6. The method ofclaim 3, wherein the output is selected from the plurality of possible outputs based on a predetermined order of the plurality of possible outputs.

7. The method ofclaim 1, wherein generating the output dialog responsive to the user speech further comprises obtaining information in accordance with the determined intent, wherein the one or more conditions is based on the obtained information.

8. The method ofclaim 1, wherein the one or more conditions depend on contextual information.

9. The method ofclaim 8, wherein the contextual information comprises a type of user device used to operate the virtual assistant, an operational state of the user device, a current time, a location of the user device, an orientation of the user device, or an orientation of a user with respect to the user device.

10. The method ofclaim 1, wherein each functional block of the plurality of sets of functional blocks includes a plurality of possible outputs, and wherein the plurality of possible outputs each include one or more of a string of words and a variable word.

11. The method ofclaim 1, wherein the dialog template further defines a format for displaying the output dialog.

12. The method ofclaim 1, wherein each output segment of the plurality of output segments is identified as being a spoken output segment or a printed output segment, and wherein outputting the output dialog comprises:

outputting a spoken output dialog based on output segments identified as being a spoken output segment; and

outputting a printed output dialog based on output segments identified as being printed output segments.

13. The method ofclaim 1, wherein each output segment of the plurality of output segments is generated independent of the other output segments.

14. The method ofclaim 1, wherein the method further comprises tagging, after accessing the textual representation of the user speech and before generating the output dialog responsive to the user speech, a word of the user speech with a dialog tag.

15. The method ofclaim 14, wherein the dialog tag comprises a synonym of the tagged word.

16. The method ofclaim 15, wherein generating the output dialog responsive to the user speech further comprises replacing an occurrence of the tagged word in one or more of the plurality of output segments with the synonym of the tagged word.

17. The method ofclaim 1, further comprising:

performing, speech to text conversion on the received user speech to generate the textual representation of the user speech.

18. The method ofclaim 1, further comprising:

in accordance with a determination that the one or more conditions are not satisfied, generating, with the first set of functional blocks, a second string for the first output segment of the plurality of output segments.

19. The method ofclaim 1, further comprising:

in accordance with a determination that the one or more conditions are not satisfied, forgo generating, with the first set of functional blocks, the first string for the first output segment of the plurality of output segments.

20. The method ofclaim 1, wherein a third set of functional blocks in the plurality of sets of functional blocks includes a set of candidate output segments, and wherein generating the plurality of output segments from the plurality of sets of functional blocks further comprises:

executing a first function defined in a first functional block of the third set of functional blocks to select a candidate output segment from the set of candidate output segments; and

generating one of the plurality of output segments by applying, to the selected candidate output segment, a second function defined in a second functional block of the third set of functional blocks, wherein the first functional block is a child block of the second functional block.

21. A non-transitory computer-readable storage medium comprising computer-executable instructions for:

receiving a user speech;

accessing a textual representation of the user speech;

determining an intent corresponding to the textual representation of the user speech;

generating an output dialog responsive to the user speech, comprising:

selecting, in accordance with the determined intent, a dialog template from a plurality of dialog templates, wherein the dialog template includes a plurality of sets of functional blocks, wherein a first set of functional blocks of the plurality of sets of functional blocks defines one or more conditions;

generating a plurality of output segments from the plurality of sets of functional blocks, including:

determining whether the one or more conditions are satisfied; and

in accordance with a determination that the one or more conditions are satisfied, generating, with the first set of functional blocks, a first string for a first output segment of the plurality of output segments;

concatenating the plurality of output segments according to an order defined by the dialog template; and

outputting the output dialog comprising the concatenated plurality of output segments.

22. The non-transitory computer-readable storage medium ofclaim 21, further comprising computer executable instructions for:

identifying a task flow based on the determined user intent; and

executing the identified task flow.

23. The non-transitory computer-readable storage medium ofclaim 21, wherein generating the plurality of output segments from the plurality of sets of functional blocks further comprises:

selecting an output from a plurality of possible outputs for a second set of functional blocks of the plurality of sets of functional blocks.

24. The non-transitory computer-readable storage medium ofclaim 21, wherein the output is selected randomly from the plurality of possible outputs.

25. The non-transitory computer-readable storage medium ofclaim 21, further comprising computer executable instructions for:

in accordance with a determination that the one or more conditions are not satisfied, generating, with the first set of functional blocks, a second string for the first output segment of the plurality of output segments.

26. The non-transitory computer-readable storage medium ofclaim 21, further comprising computer executable instructions for:

in accordance with a determination that the one or more conditions are not satisfied, forgo generating, with the first set of functional blocks, the first string for the first output segment of the plurality of output segments.

27. The non-transitory computer-readable storage medium ofclaim 21, wherein a third set of functional blocks in the plurality of sets of functional blocks includes a set of candidate output segments, and wherein generating the plurality of output segments from the plurality of sets of functional blocks further comprises:

executing a first function defined in a first functional block of the third set of functional blocks to select a candidate output segment from the set of candidate output segments; and

generating one of the plurality of output segments by applying, to the selected candidate output segment, a second function defined in a second functional block of the third set of functional blocks, wherein the first functional block is a child block of the second functional block.

28. A system comprising:

one or more processors;

memory;

one or more programs, wherein the one or more programs are stored in the memory and configured to be executed by the one or more processors, the one or more programs including instructions for:

receiving a user speech;

accessing a textual representation of the user speech;

determining an intent corresponding to the textual representation of the user speech;

generating an output dialog responsive to the user speech, comprising:

selecting, in accordance with the determined intent, a dialog template from a plurality of dialog templates, wherein the dialog template includes a plurality of sets of functional blocks, wherein a first set of functional blocks of the plurality of sets of functional blocks defines one or more conditions;

generating a plurality of output segments from the plurality of sets of functional blocks, including:

determining whether the one or more conditions are satisfied; and

in accordance with a determination that the one or more conditions are satisfied, generating, with the first set of functional blocks, a first string for a first output segment of the plurality of output segments

concatenating the plurality of output segments according to an order defined by the dialog template; and

outputting the output dialog comprising the concatenated plurality of output segments.

29. The system ofclaim 28, wherein the one or more programs further include instructions for:

identifying a task flow based on the determined user intent; and

executing the identified task flow.

30. The system ofclaim 28, wherein generating the plurality of output segments from the plurality of sets of functional blocks further comprises:

selecting an output from a plurality of possible outputs for a second set of functional blocks of the plurality of sets of functional blocks.

31. The system ofclaim 30, wherein the output is selected randomly from the plurality of possible outputs.

32. The system ofclaim 31, wherein each of the plurality of possible outputs is associated with a weighting factor, and wherein the output is selected randomly from the plurality of possible outputs based on the weighting factors.

33. The system ofclaim 30, wherein the output is selected from the plurality of possible outputs based on a predetermined order of the plurality of possible outputs.

34. The system ofclaim 28, wherein the one or more programs further include instructions for:

in accordance with a determination that the one or more conditions are not satisfied, generating, with the first set of functional blocks, a second string for the first output segment of the plurality of output segments.

35. The system ofclaim 28, wherein the one or more programs further include instructions for:

in accordance with a determination that the one or more conditions are not satisfied, forgo generating, with the first set of functional blocks, the first string for the first output segment of the plurality of output segments.

36. The system ofclaim 28, wherein a third set of functional blocks in the plurality of sets of functional blocks includes a set of candidate output segments, and wherein generating the plurality of output segments from the plurality of sets of functional blocks further comprises:

executing a first function defined in a first functional block of the third set of functional blocks to select a candidate output segment from the set of candidate output segments; and

generating one of the plurality of output segments by applying, to the selected candidate output segment, a second function defined in a second functional block of the third set of functional blocks, wherein the first functional block is a child block of the second functional block.

Images